Dynamic collaborative optimization control method for denitrification in waste incinerator

Abstract

The embodiment of the present invention discloses a method for dynamic collaborative optimization control of denitrification of a waste incinerator, which relates to the technical field of waste incineration treatment. The method comprises: obtaining the real-time collected operating parameters of the waste incinerator and performing preprocessing; based on the mechanism of waste incineration and denitrification reaction, combined with historical operating data, establishing a dynamic model of the denitrification process of the waste incinerator, wherein an autoregressive sliding average model is used to describe the denitrification process; according to the preprocessed operating parameters and the established dynamic model, the working conditions of the waste incinerator are identified and classified in real time; a multivariable model predictive control is used to formulate a collaborative control strategy; and the formulated collaborative control strategy is output to the actuators of the denitrification system and the incineration system. The method of the present invention can improve the denitrification efficiency, accurately control nitrogen oxide emissions, and reduce the ammonia escape rate and denitrification costs.

Description

Technical Field

[0001] The present invention relates to the technical field of waste incineration treatment, and in particular to a dynamic coordinated optimization control method for denitrification of a waste incinerator. Background Art

[0002] With the acceleration of urbanization, garbage disposal has gradually become a focus of social attention. As an effective garbage disposal method, garbage incineration has been widely used around the world. Garbage incineration not only reduces the volume of garbage and alleviates landfill pressure, but also generates heat energy to supply urban thermal systems. However, during the garbage incineration process, combustion efficiency and pollution emissions have always been key challenges in the fields of technology and environmental protection. Parameters such as temperature, pressure, flue gas composition and oxygen content in the incinerator have a significant impact on the efficiency and emission levels of the combustion process. If these parameters cannot be monitored and accurately adjusted in real time, it may lead to low combustion efficiency and excessive pollutant emissions, affecting environmental protection and energy utilization.

[0003] Invention patent application CN107890770A discloses an SNCR acoustic temperature measurement and zoned injection system, the technical highlights of which include:

[0004] Acoustic temperature measurement: The temperature of the furnace cross section is measured by an acoustic wave sensor, a two-dimensional temperature distribution map is constructed, and the spray gun injection is controlled in different zones.

[0005] Temperature window matching: Dynamically select the spray gun to be put into / out according to the optimal reaction temperature range of the reducing agent (ammonia / urea) (such as ammonia 850-950℃).

[0006] This invention patent application has the following disadvantages: it is only applicable to the SNCR process, and relies on the accuracy of the acoustic wave sensor. High temperature and high dust environment may affect the temperature measurement reliability.

[0007] Invention patent CN113578006B discloses an SCR denitrification control method based on control strategy optimization. Its technical key points include:

[0008] Cascade PID control: The main controller stabilizes the outlet NOx concentration, the sub-controller optimizes the ammonia injection amount, and the feedforward effect is combined to improve dynamic response.

[0009] Variable parameter design: PID parameters are adjusted according to the rate of change of inlet NOx concentration, and anti-interference design is used to deal with data distortion during CEMS instrument maintenance.

[0010] This invention patent has the following disadvantages: (1) It relies on a linearized model and has limited adaptability to extreme working conditions (such as drastic fluctuations in garbage composition); (2) The feedforward parameters require manual experience to adjust, and the degree of intelligence is low.

[0011] Invention patent application CN119532742A discloses a waste incinerator combustion management system, the technical features of which include:

[0012] Data analysis applications of the ARMA model: The Autoregressive Moving Average (ARMA) model is used to analyze the cyclical fluctuations in combustion data from waste incinerators. By building a time series model of historical combustion parameters (such as temperature and oxygen content), the cyclical fluctuation patterns of the combustion process are identified. The core of this approach is to optimize the model order using the AIC / BIC criteria to capture the trend and seasonal characteristics of the combustion data.

[0013] Modular design of the combustion management system: The system includes three modules: data acquisition, ARMA analysis, and combustion parameter adjustment. The ARMA model output guides the adjustment of parameters such as the burner damper opening and feed rate to achieve preliminary optimization of combustion efficiency.

[0014] The invention patent application has the following shortcomings: (1) Lack of closed-loop control mechanism. The ARMA model is only used as an offline data analysis tool and does not form a closed loop with the real-time control strategy. The model analysis results cannot automatically trigger control instructions, which will lead to control lag; (2) Single application scenario. The technology only optimizes the stability of the combustion system itself and does not involve the coordinated control of the denitrification system. It cannot solve the coupling problem of nitrogen oxide generation and denitrification efficiency during incineration; (3) Model static problem. The ARMA model parameters are fixed after training. The system characteristics changes caused by factors such as garbage composition fluctuations and equipment aging are not considered. Model mismatch is prone to occur in long-term operation. Summary of the Invention

[0015] In view of this, an embodiment of the present invention provides a dynamic collaborative optimization control method for denitrification of a waste incinerator to improve denitrification efficiency, accurately control nitrogen oxide emissions, and reduce ammonia escape rate and denitrification costs.

[0016] A dynamic collaborative optimization control method for denitrification of a garbage incinerator, comprising:

[0017] Step S101: obtaining the real-time collected operating parameters of the waste incinerator and performing pre-processing;

[0018] Step S102: Based on the mechanism of waste incineration and denitrification reaction and combined with historical operation data, a dynamic model of the denitrification process of the waste incinerator is established, wherein an autoregressive moving average model is used to describe the denitrification process;

[0019] Step S103: Based on the pre-processed operating parameters and the established dynamic model, the operating conditions of the waste incinerator are identified and classified in real time;

[0020] Step S104: using multivariable model predictive control to formulate a collaborative control strategy;

[0021] Step S105: Output the formulated coordinated control strategy to the actuators of the denitrification system and the incineration system.

[0022] Preferably, in step S101, the operating parameters include garbage feed amount, garbage composition, incinerator temperature, flue gas flow, nitrogen oxide concentration in flue gas and / or ammonia escape rate.

[0023] Preferably, in step S102, the discrete time model of the autoregressive moving average model is:

[0024] ;

[0025] Where y(t) is the system output at time t; is the autoregressive coefficient; is the sliding mean coefficient; is a white noise sequence; p and q are the orders of autoregression and moving average, respectively, which are determined using the information criterion method.

[0026] Preferably, in step S102, the method for establishing and optimizing the dynamic model includes:

[0027] Step A1: Determine the input and output variables based on the denitrification reaction mechanism of waste incineration;

[0028] Step A2: Collect at least 1000 sets of operating data under different working conditions;

[0029] Step A3: Constructing a discrete-time model , initialize the autoregressive order p and the sliding average order q;

[0030] Step A4: Screen the optimal p / q using the AIC / BIC information criterion;

[0031] Step A5: Train model parameters using historical data 、 , establish a dynamic mapping relationship between input variables and output variables;

[0032] Step A6: Evaluate the model accuracy using the root mean square error and mean absolute error. If the accuracy meets the requirements, solidify the model. If not, return to step A4.

[0033] Preferably, in step S103, the operating conditions of the waste incinerator include a stable operating condition, a variable load operating condition and a waste composition fluctuation operating condition.

[0034] Preferably, in step S103, the operating condition characteristics are described by constructing a membership function, with the load change rate and temperature fluctuation rate For example, the stable operating condition membership function is defined as:

[0035] ;

[0036] in, is the load change rate threshold, is the temperature fluctuation rate threshold; when the calculated When the calculated value is greater than or equal to the first preset threshold, it is determined that the waste incinerator is in a stable operating state; When the load change rate is less than the first preset threshold, the load change rate is greater than the second preset threshold, and the temperature fluctuation rate is less than or equal to the third preset threshold, it is determined that the waste incinerator is in a variable load condition; when the calculated When the load change rate is less than the first preset threshold, the temperature fluctuation rate is less than or equal to the second preset threshold, and the temperature fluctuation rate is greater than the third preset threshold, it is determined that the waste incinerator is in a waste composition fluctuation operating condition.

[0037] Preferably, in step S104, the optimization objective function of the multivariable model predictive control is:

[0038] ;

[0039] in, Measure the system output y(k|t) at time k predicted by time t and the target output vector y r The error between them is weighted by the weight matrix Q on the errors of different output variables; Used to constrain the size of the control input u(k|t), the weight matrix R determines the degree of penalty for changes in the control input; The deviation between the system output and the target output at the end point Np of the prediction time domain is further constrained, and the weight matrix F ensures the long-term performance of the system within the entire prediction time domain;

[0040] The constraints are:

[0041] ;

[0042] Among them, A, B, and E are the linearized system matrices, which respectively reflect the influence of the system state, control input, and external disturbance on the system state at the next moment; x(k|t) represents the prediction of the system state at time k at time t, u(k|t) is the control input at the corresponding moment, and d(k|t) represents the external disturbance; u min and u max are the lower and upper limit vectors of the control input respectively; y NOx,max and y NOx,min They are the maximum and minimum allowable concentrations of nitrogen oxide emissions, respectively.

[0043] Preferably, in step S104, the recursive least squares method is used to update the model parameters, and the parameter update law is:

[0044] ;

[0045] in, For the forgetting factor, is the regression vector, is the model parameter vector estimated in real time, and P(t) is the covariance matrix.

[0046] Preferably, the step S105 is further as follows:

[0047] The developed collaborative control strategy is output to the actuators of the denitrification system and incineration system. At the same time, the control effect is monitored in real time, the actual operation data is compared with the control target, the deviation is calculated, and the control strategy is adjusted and optimized online based on the deviation size and change trend using the PID feedback control algorithm to form a closed-loop control.

[0048] Preferably, the step S105 includes:

[0049] Step B1: Send the parameters generated by the collaborative control strategy to the execution mechanism;

[0050] Step B2: adjusting the reducing agent injection amount and the incinerator operating parameters;

[0051] Step B3: Using sensors to collect real-time information on outlet NOx concentration, ammonia escape rate, and actual reducing agent usage;

[0052] Step B4: Compare the actual value with the target value and calculate the NOx concentration deviation and the ammonia escape rate deviation;

[0053] Step B5: Preset the deviation threshold. If it does not exceed the deviation threshold, the existing control parameters are maintained and the process ends. If it exceeds the deviation threshold, go to step B6;

[0054] Step B6: Adaptively adjust the PID controller parameters according to the deviation size and change trend, fine-tune the collaborative control strategy, and go to step B1.

[0055] The dynamic collaborative optimization control method for denitrification of a garbage incinerator of the present invention has the following beneficial effects:

[0056] 1. Strong dynamic adaptability: It can perceive the changes in working conditions during the waste incineration process in real time. By establishing dynamic models and identifying and classifying working conditions, it can adjust the control strategy in a timely manner to adapt to complex working conditions such as fluctuations in waste composition and load changes, thereby improving the robustness of the denitrification system.

[0057] 2. High denitrification efficiency: Through the coordinated control and multi-variable coordinated control of the denitrification system and the incineration system, precise regulation of the denitrification process is achieved, which can effectively improve the nitrogen oxide removal rate and ensure that nitrogen oxide emissions are stable and meet standards.

[0058] 3. Reduce ammonia escape rate: Accurately control the injection amount and reaction conditions of the denitrifier, reduce excessive use of the denitrifier, reduce the ammonia escape rate, and reduce harm to the environment and subsequent equipment.

[0059] 4. Cost savings: Optimize denitrification agent and energy consumption, taking into account both environmental protection and economic benefits. Optimize the use of denitrification agent, while improving the operating efficiency of the incineration system, reducing energy consumption and denitrification costs, and achieving good economic and environmental benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0061] Figure 1 Schematic diagram of the process of the dynamic coordinated optimization control method for denitrification of a waste incinerator according to the present invention;

[0062] Figure 2 This is an overall principle diagram of the dynamic collaborative optimization control method for denitrification of a waste incinerator according to the present invention;

[0063] Figure 3 A flow chart for establishing and optimizing the dynamic model in the present invention;

[0064] Figure 4 This is a flow chart of the control strategy implementation and feedback adjustment in the present invention. DETAILED DESCRIPTION

[0065] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0066] It should be understood that the embodiments described are only a portion of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative work are within the scope of protection of the present invention.

[0067] The embodiment of the present invention provides a method for dynamic coordinated optimization control of denitrification in a garbage incinerator, such as Figure 1-2 Shown, including:

[0068] Step S101: obtaining the real-time collected operating parameters of the waste incinerator and performing pre-processing;

[0069] This step corresponds to Figure 2 "Data Collection and Processing" in.

[0070] In practice, sensors can be used to collect real-time incinerator operating parameters, including waste feed volume, waste composition (e.g., carbon, hydrogen, oxygen, nitrogen, and other elemental content), incinerator temperature (including furnace temperature and flue gas temperature), flue gas flow rate, flue gas nitrogen oxide concentration, and ammonia escape rate. The collected data is then preprocessed, including filtering, noise reduction, and normalization, to remove interference and outliers, thereby improving data accuracy and reliability.

[0071] Step S102: Based on the mechanism of waste incineration and denitrification reaction and combined with historical operation data, a dynamic model of the denitrification process of the waste incinerator is established, wherein an autoregressive moving average model is used to describe the denitrification process;

[0072] This step corresponds to Figure 2 "Building a Dynamic Model" in [1].

[0073] The model uses garbage composition, incineration temperature, flue gas flow rate, etc. as input variables, and nitrogen oxide removal rate, ammonia escape rate, etc. as output variables, and can reflect the dynamic relationship between various variables in the denitrification process.

[0074] This scheme uses the autoregressive moving average model (ARMA) to describe the denitrification process. Its discrete time model is as follows:

[0075] ;

[0076] Where y(t) is the system output at time t, such as the NOx concentration after denitrification;

[0077] is the autoregressive coefficient, which is used to characterize the impact of the system's historical output on the current output;

[0078] is the sliding average coefficient, which reflects the effect of past white noise on the current output;

[0079] is a white noise sequence, representing unpredictable random interference;

[0080] p and q represent the orders of the autoregressive and moving average, respectively, and must be determined based on the actual data characteristics. To accurately adapt to the complex characteristics of the denitrification process in waste incinerators, we used information criteria such as the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC) to determine the orders of p and q. By analyzing and calculating a large amount of actual operating data, with the goal of minimizing the AIC or BIC value, we screened for the optimal combination of p and q. This ensured that the ARMA model accurately captured the dynamic changes in the denitrification process and accurately described the time-varying characteristics of the denitrification process.

[0081] Unlike patent application CN119532742A, in which ARMA is only used for data analysis, this case combines mechanism analysis with data-driven development, and for the first time applies the ARMA model to the dynamic modeling of the denitrification process. It establishes a dynamic mapping relationship between input variables (waste composition, incineration temperature, flue gas flow, etc.) and output variables (nitrogen oxide removal rate, ammonia escape rate, etc.), solving the problem of modeling the time-varying nonlinear characteristics of the denitrification reaction.

[0082] As an optional embodiment, Figure 3 As shown, in step S102, the method for establishing and optimizing the dynamic model includes:

[0083] Step A1 (Mechanism Analysis): Based on the denitrification reaction mechanism of waste incineration (such as ammonia nitrogen redox reaction), determine the input variables (feed rate, temperature, reducing agent dosage) and output variables (NOx removal rate, ammonia escape rate).

[0084] Step A2 (Historical Data Collection): Collect at least 1,000 sets of operating data under different operating conditions, including stable load, variable load, component fluctuation, and other scenarios.

[0085] Step A3 (ARMA model initialization): Constructing a discrete-time model , initialize the autoregressive order p and the sliding average order q.

[0086] Step A4 (order optimization): Use the AIC / BIC information criterion to select the optimal p / q (e.g., p=2, q=1 in a certain case) to minimize the model prediction error.

[0087] Step A5 (Model Training): Use historical data to train model parameters 、 , establish a dynamic mapping relationship between input variables and output variables.

[0088] Step A6 (error verification): The model accuracy was evaluated by the root mean square error (RMSE) and mean absolute error (MAE). For example, the RMSE before implementation was 30 mg / Nm³, which was reduced to 10 mg / Nm³ after optimization.

[0089] Step A7 (Model Solidification): After the error meets the standard, the model is embedded in the control system for real-time prediction and control strategy generation. If the error does not meet the standard, return to step A4, that is, re-screen the optimal p / q order using the AIC / BIC criterion, adjust the autoregressive order p and the sliding average order q, and then execute steps A5-A6 again until the accuracy meets the standard and the model is solidified.

[0090] Step S103: Based on the pre-processed operating parameters and the established dynamic model, the operating conditions of the waste incinerator are identified and classified in real time;

[0091] This step corresponds to Figure 2 “Working Condition Identification and Classification” in.

[0092] As an optional embodiment, in step S103, the operating conditions of the waste incinerator include different types, such as stable operating conditions, variable load operating conditions, and waste composition fluctuation conditions. For different operating conditions, corresponding control strategies are formulated to achieve dynamic switching of control strategies.

[0093] In the method of the present invention, the construction of membership function plays a key role in accurately describing the working condition characteristics. and temperature fluctuation rate Taking these two important parameters that can significantly reflect the operating status of the waste incinerator as an example, the stable operating condition membership function is explained in detail. The stable operating condition membership function is defined as:

[0094] ;

[0095] in, is the load change rate threshold, The temperature fluctuation rate threshold is determined by analyzing historical data. The historical data covers the load changes and temperature fluctuations of the waste incinerator in different operating stages, different types of waste and different environmental conditions. After screening, sorting and statistical calculation of these data, a reasonable and value to ensure that the membership function can accurately reflect the actual working conditions and solve the problem of automatic switching of control strategies under complex working conditions (the existing technology has no multi-dimensional working condition classification).

[0096] In practical applications, when the calculated When , the system will determine that the waste incinerator is in a stable operating condition. This determination means that the load and temperature changes of the incinerator are in a relatively stable state. At this time, denitrification control can be carried out according to the strategy under conventional stable operating conditions. On the contrary, when When (load change rate And the temperature fluctuation rate For variable load conditions; load change rate And the temperature fluctuation rate (the operating condition is fluctuating due to garbage composition), the system will immediately trigger the variable operating condition control logic to cope with the instability of the incinerator's operating state, ensuring that the denitrification process can adapt to changes in operating conditions and proceed continuously and efficiently.

[0097] This function couples load and temperature fluctuations for analysis for the first time. Compared with several existing technologies mentioned above, it can more accurately identify complex working conditions (such as coupled scenarios of sudden load changes and abnormal temperature), providing a key basis for dynamic switching of control strategies.

[0098] Step S104: using multivariable model predictive control (MPC) to formulate a collaborative control strategy;

[0099] This step corresponds to Figure 2 Collaborative Control Strategy Formulation.

[0100] Coordinated control of denitrification system and incineration system:

[0101] Fluctuations in incineration system parameters, such as combustion temperature and feed rate, directly impact denitrification performance, necessitating coordinated control to dynamically match the two systems. To achieve this coordinated control, this method builds upon the advanced framework of multivariable model predictive control. This framework utilizes a rolling optimization algorithm to dynamically and precisely adjust control inputs.

[0102] Taking actual operational scenarios as an example, in an incineration system, when the amount of waste feed changes due to factors such as the source and quality of the waste, the temperature in the combustion chamber fluctuates accordingly. Simultaneously, the amount of flue gas produced will also show a corresponding increase or decrease. These changes are further transmitted to the denitrification system, directly affecting the generation of nitrogen oxides (NOx). With the help of this collaborative control strategy, the system can capture real-time state changes in the incineration system and adjust the key control parameters of the denitrification system in advance. This ensures that NOx emissions meet environmental standards while allowing the incineration system to maintain efficient and stable operation, achieving a win-win situation for environmental protection and production efficiency.

[0103] Multivariable coordinated control:

[0104] This invention uses multivariable model predictive control to coordinate the relationships between multiple variables. Within the prediction horizon Np, it is dedicated to solving a rolling optimization problem. The core of this problem is to rationally adjust the control inputs so that the system output approaches the desired target as closely as possible, while ensuring that the changes in the control inputs remain within a reasonable range to avoid over-adjustment that may lead to system instability or increased energy consumption. The optimization objective function can be expressed as:

[0105] ;

[0106] This objective function comprehensively considers the deviation between the system output and the target output in the next Np time steps from the current time t, as well as the size of the control input. It measures the system output y(k|t) predicted at time k and the target output vector y r The errors between the output variables are weighted using the weight matrix Q to highlight the output indicators of key interest. For example, in a waste incineration scenario, if the control accuracy of nitrogen oxide emission concentration is crucial, the elements related to nitrogen oxide emissions can be given a larger weight in the weight matrix Q, so that the system prioritizes reducing the error of this indicator during the optimization process. It is used to constrain the size of the control input u(k|t). The weight matrix R determines the degree of penalty for changes in the control input to avoid excessive control input leading to system instability or additional energy consumption.

[0107] In actual equipment operation, excessive control inputs can subject equipment components to excessive stress or current, shortening their lifespan and causing unnecessary energy waste. By setting an appropriate weight matrix, R, we can effectively limit the amplitude of control inputs, ensuring that the system operates within a safe and energy-efficient range.

[0108] and The weight matrix F further constrains the deviation between the system output and the target output at the end point Np of the prediction time domain. This ensures the long-term performance of the system throughout the prediction time domain and enhances closed-loop stability. This serves as a clear target boundary for the system's long-term operation, preventing it from gradually deviating from its intended trajectory and ensuring that it consistently delivers results that meet expectations over time.

[0109] To ensure the feasibility and practical physical significance of the optimization problem, the optimization process must meet a series of strict constraints, as follows:

[0110] ;

[0111] Among them, the state transfer equation Describes the evolution of the system state over time.

[0112] A, B, and E are the linearized system matrices, which respectively reflect the influence of the system state, control input, and external disturbance on the system state at the next moment.

[0113] x(k|t) represents the predicted system state at time k at time t, u(k|t) is the control input at that time, and d(k|t) represents the external disturbance. For example, in a waste incinerator, changes in ambient temperature or sudden changes in waste composition can be considered external disturbances d(k|t). Using this equation, the system can accurately predict the state at the next moment based on the current state, control input, and external disturbances, providing a solid data foundation for subsequent control decisions.

[0114] Control input constraint u min ≤u(k|t)≤u max This ensures that the control input is always within the physically achievable range, preventing damage to the device or malfunction due to excessive or insufficient control input. min and u max are the lower and upper limit vectors of the control input, and each element corresponds to the value range of a control variable. Taking the denitrification agent injection pump as an example, its working flow has a safe operating range, and the corresponding flow value is the lower limit u of the control input variable. min and upper limit u max Any control instruction beyond this range may cause equipment failure or failure to effectively perform the denitrification task.

[0115] For the key indicator y in the system output NOx (For example, the concentration of nitrogen oxides emitted during the denitrification process of a waste incinerator) also has strict upper and lower limits. .

[0116] y NOx,max and y NOx,min The maximum and minimum values ​​of nitrogen oxide emission concentration are allowed respectively. This constraint ensures that the system always meets the relevant requirements of environmental protection during operation and controls nitrogen oxide emissions within a reasonable range. Environmental protection laws and regulations have clear and strict standards for nitrogen oxide emissions from waste incineration plants. Once the emission concentration exceeds the upper limit y NOx,max , will face the risk of high fines or even suspension of production for rectification; and if the emission concentration is far below the lower limit for a long time NOx,min Although it meets environmental protection requirements, it may mean that the denitrification process consumes excessive resources and affects production efficiency.

[0117] During actual operation, system model parameters may change due to various factors (such as equipment aging and environmental changes), thus affecting control effectiveness. To address this issue, this method uses recursive least squares (RLS) to update model parameters online, tracking changes in system characteristics in real time to ensure the effectiveness and adaptability of the control algorithm. The parameter update law is as follows:

[0118] ;

[0119] in, is the forgetting factor, which usually ranges from [0,1] and is used to adjust the forgetting speed of historical data. A small value means that the algorithm pays more attention to recent data and responds faster to system changes, but it may be more sensitive to noise. For example, when the area where the waste incinerator is located suddenly encounters extreme weather and the equipment operating environment changes drastically, a small value will cause the algorithm to pay more attention to recent data and respond faster to system changes, but it may be more sensitive to noise. It enables the algorithm to quickly capture these changes and adjust the model parameters in time to adapt to the new working conditions. However, if there is a lot of noise interference in the system measurement data, the smaller A value of 0 may cause parameter updates to be affected by noise and experience large fluctuations. The value of makes the algorithm more dependent on historical data, which can smooth the noise influence to a certain extent, but has a relatively weak ability to track the dynamic changes of the system. The value helps to utilize the effective information in historical data and improve the stability and accuracy of parameter estimation. , which can strike a balance between tracking system changes and suppressing noise.

[0120] is a regression vector, which contains the historical input and output information related to the system output, and uses this information to build the relationship between the model parameters and the system output. It may cover data such as the amount of garbage feed, combustion temperature, denitrification agent injection volume and corresponding nitrogen oxide emission concentration over a period of time in the past. Through the combination of regression vectors, it provides a rich information basis for the algorithm to accurately estimate model parameters.

[0121] T represents the matrix transpose operator (Transpose), which is used to transform the regression vector Convert from row vectors to column vectors to ensure dimensionality consistency for matrix multiplication.

[0122] The model parameter vector is estimated in real time. As new data is continuously input, the algorithm uses a recursive formula to continuously update the parameter vector so that it can accurately reflect the current characteristics of the system. For example, as a waste incinerator runs for a long time, the refractory material in the furnace gradually wears out, the combustion efficiency changes, and the model parameters will also change accordingly. Through the recursive least squares method, These changes can be tracked in a timely manner to ensure that the model always matches the actual system.

[0123] P(t) is the covariance matrix, which measures the uncertainty of parameter estimates. Each time the parameters are updated, the covariance matrix is ​​adjusted accordingly to reflect changes in parameter estimation accuracy as data accumulates. When the system is initially launched and the amount of data is relatively small, the covariance matrix P(t) is large, indicating high uncertainty in the parameter estimates. As the system runs longer and data accumulates, P(t) gradually decreases, and the accuracy of the parameter estimates continues to improve. Through the parameter update process described above using the recursive least squares method, the system can adapt to changes in model parameters in real time, thereby maintaining consistently good control performance.

[0124] In this step, multivariable model predictive control (MPC) is used to construct the optimization objective function for different working conditions, and rolling optimization is performed in combination with equipment constraints and environmental protection requirements. The model parameters are updated online using the RLS algorithm.

[0125] Unlike patent application CN118836453A, in which MPC is only used for single-system (combustion-supporting air) control, the present invention is the first to extend the MPC algorithm to the coordinated control of the denitrification and incineration dual systems. By constructing an optimization objective function that includes cross-system constraints (such as the incineration temperature affecting the denitrification reaction efficiency), dynamic matching of the dual systems is achieved, solving the problem of inter-system regulation lag in the existing technology.

[0126] This invention realizes the online adaptive update of ARMA model parameters for the first time (different from the existing technologies mentioned above). Dynamically adjusting the weights of historical data enables the model to track changes in system characteristics caused by fluctuations in garbage composition and equipment aging in real time, filling the problem of decreased control accuracy caused by model mismatch in existing technologies.

[0127] Step S105: Output the formulated coordinated control strategy to the actuators of the denitrification system and the incineration system.

[0128] This step corresponds to Figure 2 "Control strategy implementation and feedback adjustment" in.

[0129] Preferably, this step further includes: outputting the developed collaborative control strategy to the actuators of the denitration system and the incineration system, such as the denitrification agent injection pump and the incinerator burner, to achieve real-time control of the denitration and incineration processes. At the same time, the control effect is monitored in real time, the actual operating data is compared with the control target, the deviation is calculated, and the control strategy is adjusted and optimized online based on the deviation size and change trend using the PID feedback control algorithm to form a closed-loop control to ensure that the denitration system is always in the optimal operating state.

[0130] In this way, a complete closed loop of "dynamic modeling-operating condition identification-coordinated control-real-time optimization" is formed. Different from the single control logic of existing technologies (such as patent application CN119532742A which only has one-way control and patent application CN118836453A which has no parameter adaptation), the real-time adaptation / self-optimization of the model and control strategy is achieved through the coordination of PID feedback and RLS update, ensuring long-term stable operation under complex operating conditions.

[0131] Preferably, Figure 4 As shown, step S105 includes:

[0132] Step B1 (control instruction output): Send the parameters generated by the collaborative control strategy (such as ammonia injection amount and combustion temperature setting value) to the actuator (injection pump, burner).

[0133] Step B2 (actuator action): adjust the reducing agent injection amount (such as dynamically adjusting from 500L / h to 650L / h) and the incinerator operating parameters (such as furnace temperature 850℃±50℃).

[0134] Step B3 (Real-time Monitoring): Sensors are used to collect data such as outlet NOx concentration (target ≤ 100 mg / Nm³), ammonia escape rate (target ≤ 7.8 ppm), and actual reducing agent usage in real time.

[0135] Step B4 (deviation calculation): Compare the actual value with the target value and calculate the NOx concentration deviation (such as the actual , deviation 5mg / Nm³) and ammonia escape rate deviation.

[0136] Step B5 (deviation determination): A preset deviation threshold (such as NOx concentration fluctuation > 6.5% or ammonia escape rate > 1.5 ppm) is set. If the threshold is exceeded, feedback adjustment is triggered.

[0137] Step B6 (PID parameter adjustment): According to the deviation size and change trend, adaptively adjust the PID controller parameters (such as Kp=0.8 for stable working conditions and Kp=1.2 for variable working conditions), fine-tune the control strategy, and go to step B1.

[0138] Step B7 (maintain current strategy): When the deviation is within the allowable range, maintain the existing control parameters to avoid excessive adjustment that may cause system fluctuations. The process ends.

[0139] Implementation Cases

[0140] 1) Data collection and processing

[0141] Taking a 750t / d domestic waste incineration power plant in Zhejiang as an example, the following core parameters are collected in real time:

[0142] Waste feed rate: 750t / d (fluctuation ±5% at stable load);

[0143] Inlet NOx concentration: 300mg / Nm³ (standard state, 11% O2), outlet index daily average value ≤100mg / Nm³, current process detection value is 150mg / Nm³;

[0144] Reducing agent: 20% ammonia water, initial injection volume 500L / h (corresponding to a NOx removal rate of approximately 50%);

[0145] Incineration temperature: furnace center temperature 850±50℃, flue gas temperature 300±30℃;

[0146] Flue gas flow rate: 120,000Nm³ / h (standard).

[0147] 2) Build a dynamic model

[0148] ARMA model parameter calculation:

[0149] The outlet NOx concentration is used as the output variable y(t), and the input variables include the feed rate L(t), the inlet NOx concentration CNOx,in(t), and the ammonia injection rate u(t). The optimal order p=2, q=1 is determined through historical data training, and the model expression is: .

[0150] Model Validation:

[0151] Prediction error before implementation: RMSE=30mg / Nm³, MAE=25mg / Nm³;

[0152] After implementation, through RLS online update, RMSE was reduced to 10mg / Nm³ and MAE=8mg / Nm³, meeting the forecast accuracy requirement of export index ≤100mg / Nm³.

[0153] 3) Working condition identification and classification Stable working condition determination:

[0154] Load change rate ,when And the temperature fluctuation rate When , the membership function is calculated as: , it is determined to be a stable working condition and the conventional control strategy is triggered; if the feed rate fluctuation is greater than 5% (such as an instantaneous increase to 800t / d), then , start the variable load control logic.

[0155] 4) Collaborative control strategy formulation and multivariable optimization calculation:

[0156] The objective function is to optimize the ammonia injection amount u(t) with the outlet NOx≤100mg / Nm³ as the constraint. p=10, weight matrix Q=diag[10,1] (NOx weight priority), R=diag[0.1] (limit injection quantity fluctuation).

[0157] Constraints: State transfer equation: (Simplified model);

[0158] Control input: 200L / h≤u(k)≤800L / h (safety range of ammonia injection pump);

[0159] Output constraint: y NOx,out ≤100mg / Nm³.

[0160] RLS parameter update: forgetting factor When the inlet NOx concentration suddenly rises to 350mg / Nm³, the algorithm completes the parameter update within 5 time steps (about 30 seconds), and the injection volume is dynamically adjusted from 500L / h to 650L / h, ensuring that the outlet concentration is stable at 95±5mg / Nm³.

[0161] The comparison of data improvement before and after implementing the method of the present invention is shown in Table 1:

[0162] Table 1

[0163] index Before implementation After implementation Improvement Denitrification efficiency 52.1% 66.7% 28% Ammonia escape rate 11.3ppm ≤7.8ppm 31% Reducing agent dosage 520L / h 425L / h 18.3% Load response time 10 minutes 3 minutes 70% Standard stability Daily average fluctuation ±18.3% Daily average fluctuation is ±6.7% 2.7 times improvement

[0164] The method of this invention utilizes ARMA, MPC, and RLS technologies not simply as a superposition, but rather achieves deep collaboration through the operating condition identification module. ARMA provides precise predictions for MPC, and MPC outputs drive RLS to update model parameters online, forming a closed "prediction-control-optimization" loop. This systematic innovation, not addressed in the aforementioned existing technologies, effectively addresses control lag and parameter mismatch issues under complex operating conditions.

[0165] The method of the present invention realizes the coordinated control of the denitrification and incineration systems for the first time, and synchronously adjusts the incineration temperature and the denitrification agent dosage through a multivariable optimization objective function (for example, when the nitrogen content of the garbage suddenly increases, the incineration temperature is automatically coordinated to increase to reduce NOx generation, and the reducing agent injection amount is optimized at the same time). This is different from the existing patent CN118836453A, which only regulates the single combustion air system.

[0166] In summary, the dynamic collaborative optimization control method for denitrification of a waste incinerator of the present invention is aimed at the problem of nitrogen oxide (NOx) emission control during the waste incineration process. It integrates technologies such as dynamic modeling, real-time identification of operating conditions, multivariable collaborative control, and autoregressive moving average model (ARMA) to achieve collaborative optimization of the denitrification system and the incineration system. It is suitable for efficient denitrification control under complex operating conditions such as fluctuations in waste composition and changes in load, and belongs to the intersection of industrial process control and environmental pollution control. The method of the present invention can dynamically adjust the denitrification control strategy according to the real-time changes in operating conditions during the waste incineration process, achieve collaborative optimization of the denitrification system and the incineration system, improve denitrification efficiency, accurately control nitrogen oxide emissions, and reduce ammonia escape rate and denitrification costs. The method also has the following beneficial effects:

[0167] 1. Strong dynamic adaptability: It can perceive the changes in working conditions during the waste incineration process in real time. By establishing dynamic models and identifying and classifying working conditions, it can adjust the control strategy in a timely manner to adapt to complex working conditions such as fluctuations in waste composition and load changes, thereby improving the robustness of the denitrification system.

[0168] 2. High denitrification efficiency: Through the coordinated control and multi-variable coordinated control of the denitrification system and the incineration system, precise regulation of the denitrification process is achieved, which can effectively improve the nitrogen oxide removal rate and ensure that nitrogen oxide emissions are stable and meet standards.

[0169] 3. Reduce ammonia escape rate: Accurately control the injection amount and reaction conditions of the denitrifier, reduce excessive use of the denitrifier, reduce the ammonia escape rate, and reduce harm to the environment and subsequent equipment.

[0170] 4. Cost savings: Optimize denitrification agent and energy consumption, taking into account both environmental protection and economic benefits. Optimize the use of denitrification agent, while improving the operating efficiency of the incineration system, reducing energy consumption and denitrification costs, and achieving good economic and environmental benefits.

[0171] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A dynamic collaborative optimization control method for denitrification of a waste incinerator, characterized in that: include: Step S101: obtaining the real-time collected operating parameters of the waste incinerator and performing pre-processing; Step S102: Based on the mechanism of waste incineration and denitrification reaction and combined with historical operation data, a dynamic model of the denitrification process of the waste incinerator is established, wherein an autoregressive moving average model is used to describe the denitrification process; Step S103: Based on the pre-processed operating parameters and the established dynamic model, the operating conditions of the waste incinerator are identified and classified in real time; Step S104: using multivariable model predictive control to formulate a collaborative control strategy; Step S105: Outputting the formulated coordinated control strategy to the actuators of the denitrification system and the incineration system; Wherein, in step S102, the discrete time model of the autoregressive moving average model is: ； Where y(t) is the system output at time t; is the autoregressive coefficient; is the sliding mean coefficient; is a white noise sequence; p and q are the orders of autoregression and moving average, respectively, determined using the information criterion method; In step S103, the operating conditions of the waste incinerator include stable operating conditions, variable load operating conditions, and waste composition fluctuation operating conditions. The operating condition characteristics are described by constructing a membership function. The stable operating condition membership function is defined as: ； in, is the load change rate, is the temperature fluctuation rate, is the load change rate threshold, is the temperature fluctuation rate threshold; when the calculated When the calculated value is greater than or equal to the first preset threshold, it is determined that the waste incinerator is in a stable operating state; When the load change rate is less than the first preset threshold, the load change rate is greater than the second preset threshold, and the temperature fluctuation rate is less than or equal to the third preset threshold, it is determined that the waste incinerator is in a variable load condition; when the calculated When the load change rate is less than the first preset threshold, the load change rate is less than or equal to the second preset threshold, and the temperature fluctuation rate is greater than the third preset threshold, it is determined that the waste incinerator is in a waste composition fluctuation operating condition; In step S104, the optimization objective function of the multivariable model predictive control is: ； in, Measure the system output y(k|t) at time k predicted by time t and the target output vector y r The error between them is weighted by the weight matrix Q on the errors of different output variables; Used to constrain the size of the control input u(k|t), the weight matrix R determines the degree of penalty for changes in the control input; The deviation between the system output and the target output at the end point Np of the prediction time domain is further constrained, and the weight matrix F ensures the long-term performance of the system within the entire prediction time domain; The constraints are: ； Among them, A, B, and E are the linearized system matrices, which respectively reflect the influence of the system state, control input, and external disturbance on the system state at the next moment; x(k|t) represents the prediction of the system state at time k at time t, u(k|t) is the control input at the corresponding moment, and d(k|t) represents the external disturbance; u min and u max are the lower and upper limit vectors of the control input respectively; y NOx,max and y NOx,min They are the maximum and minimum allowable concentrations of nitrogen oxide emissions, respectively.

2. The dynamic collaborative optimization control method for denitrification of a waste incinerator according to claim 1 is characterized in that: In step S101, the operating parameters include the amount of garbage feed, garbage composition, incinerator temperature, flue gas flow, nitrogen oxide concentration in the flue gas and / or ammonia escape rate.

3. The dynamic collaborative optimization control method for denitrification of a waste incinerator according to claim 1 is characterized in that: In step S102, the method for establishing and optimizing the dynamic model includes: Step A1: Determine the input and output variables based on the denitrification reaction mechanism of waste incineration; Step A2: Collect at least 1000 sets of operating data under different working conditions; Step A3: Constructing a discrete-time model , initialize the autoregressive order p and the sliding average order q; Step A4: Screen the optimal p / q using the AIC / BIC information criterion; Step A5: Train model parameters using historical data 、 , establish a dynamic mapping relationship between input variables and output variables; Step A6: Evaluate the model accuracy using the root mean square error and mean absolute error. If the accuracy meets the requirements, solidify the model. If not, return to step A4.

4. The method for dynamic coordinated optimization control of denitrification in a waste incinerator according to claim 1, characterized in that: In step S104, the recursive least squares method is used to update the model parameters, and the parameter update law is: ； in, For the forgetting factor, is the regression vector, is the model parameter vector estimated in real time, and P(t) is the covariance matrix.

5. The method for dynamic coordinated optimization control of denitrification in a waste incinerator according to any one of claims 1 to 4, characterized in that: The step S105 is further as follows: The developed collaborative control strategy is output to the actuators of the denitrification system and incineration system. At the same time, the control effect is monitored in real time, the actual operation data is compared with the control target, the deviation is calculated, and the control strategy is adjusted and optimized online based on the deviation size and change trend using the PID feedback control algorithm to form a closed-loop control.

6. The method for dynamic coordinated optimization control of denitrification in a waste incinerator according to claim 5, characterized in that: The step S105 includes: Step B1: Send the parameters generated by the collaborative control strategy to the execution mechanism; Step B2: adjusting the reducing agent injection amount and the incinerator operating parameters; Step B3: Using sensors to collect real-time information on outlet NOx concentration, ammonia escape rate, and actual reducing agent usage; Step B4: Compare the actual value with the target value and calculate the NOx concentration deviation and the ammonia escape rate deviation; Step B5: Preset the deviation threshold. If it does not exceed the deviation threshold, the existing control parameters are maintained and the process ends. If it exceeds the deviation threshold, go to step B6; Step B6: Adaptively adjust the PID controller parameters according to the deviation size and change trend, fine-tune the collaborative control strategy, and go to step B1.

Images